1.1  Polycystic Ovary Syndrome


Polycystic ovary syndrome (PCOS) is a heterogeneous disorder that affects approximately 6-10% of women of reproductive age (Franks, 1995). Polycystic ovary syndrome is probably the most prevalent endocrinopathy in women and by far the most common cause for infertility. In fact, polycystic ovaries have been associated with 75% of cases of anovulation (Hull, 1987).

1.2 Presentation

The many features of this syndrome can be divided into three categories: clinical, endocrine and metabolic. The clinical features include menstrual abnormalities, hirsutism, acne, alopecia, anovulatory infertility and recurrent miscarriages. The endocrine features are presented with elevated androgens, luteinizing hormone, and oestrogen and prolactin levels. The metabolic aspects of this syndrome are insulin resistance, obesity, lipid abnormalities and an increased risk for impaired glucose tolerance and type 2 diabetes mellitus (type 2 DM).

Endocrine Abnormalities

The main endocrine features of PCOS are increased androgen production and disordered gonadotropin secretion. Both luteinizing hormone (LH) pulse frequency and amplitude are increased, whereas follicle-stimulating hormone (FSH) levels remain constant in the midfollicular range (Marshall et al., 1999). The frequency of gonadotropin-releasing hormone (GnRH) release is increased secondary to decreased sensitivity of the GnRH pulse generator to the negative feedback effects of estradiol and progesterone. This increased GnRH pulse frequency selectively increases LH release. The raised LH levels enhance thecal androgen production, and these androgens are incompletely aromatised into estrogens by the granulosa cells, because of arrested follicular development as a consequence of low-level cyclic FSH release. The so-called vicious cycle of PCOS is created, in which disordered gonadotropin secretion causes increased ovarian androgen production, which in turn alters gonadal steroid feedback, perpetuating disordered gonadotropin release (Dunaif, 1997). Adrenal androgen production is also frequently increased in PCOS (Rosenfield, 1999). This finding might reflect a common defect in ovarian and adrenal androgen biosynthesis because adrenocorticotropin hormone (ACTH) release is not increased.

Metabolic Features


Insulin resistance is a prominent feature of PCOS, independent of obesity (Dunaif, 1992). Many but not all women with PCOS are insulin resistant. Obesity and PCOS have an additive deleterious effect on insulin sensitivity. The molecular mechanisms of this defect differ from those in other common insulin resistant conditions, such as Type 2 Diabetes mellitus and obesity, suggesting that PCOS-related insulin resistance has an individual genetic aetiology (Dunaif, 1997). Studies of PCOS adipocytes suggest that there is a post-binding defect in insulin receptor-mediated signal transduction, and this observation has recently been confirmed in skeletal muscle, the major site of insulin-mediated glucose uptake (Dunaif et al., 2001). Studies of insulin receptors isolated from PCOS-cultured skin fibroblasts suggested that the signalling defect results from a decrease in insulin receptor tyrosine kinase activity, secondary to a constitutive increase in receptor serine phosphorylation (Dunaif, 1997). The signalling defect produces selective insulin resistance, affecting metabolic but not mitogenic actions of insulin (Bock et al., 1999). Insulin acts through its cognate receptor, in synergy with LH, to stimulate theca cell steroidogenesis in PCOS (Nestler et al., 1998). Hence, it is possible that the selective insulin resistance of PCOS accounts for the continued actions of insulin on steroidogenesis, in spite of defects in insulin-mediated glucose metabolism. Insulin also contributes to increased adrenal androgen secretion in PCOS, in part by enhancing adrenal sensitivity to ACTH (Moghetti et al., 1996).

Ek and colleagues (2002) have identified fat depot-specific abnormalities in the regulation of lipolysis in PCOS. Isolated subcutaneous abdominal adipocytes are resistant to catecholamine-induced lipolysis in women with PCOS. The opposite phenomenon, markedly enhanced sensitivity, is observed in their visceral adipocytes (Large et al., 1998). The increase in visceral fat lipolysis could lead to an increase in free fatty acids (FFA) release that subsequently contributes to hepatic glucose production (Montague et al., 2000). Accordingly, enhanced visceral fat lipolysis could be one mechanism for the increased risk for glucose intolerance in PCOS (Bergman, 1997).

1.3 Polycystic Ovaries 

The classical ultrasound features of PCOS, which have been previously described (Adams et al., 1985) include an enlarged ovary with the presence of 10 or more cysts, 2-8 mm in diameter, arranged either peripherally around a dense core of stroma or scattered throughout an increased amount of stroma. However, up to 23% of normal women meet the sonographic criteria for polycystic ovaries. On the other hand, many investigators report that ovaries from women with PCOS may be normal (Timor-Tritsch et al., 1998) The presence of polycystic ovaries was not included in the definition of PCOS.


Figure I: Transvaginal ultrasound of the polycystic ovary

1.4 Pathogenesis Of PCOS 

A series of investigations have emphasized a heterogeneous nature of PCOS with different combinations of features present in individual patients. The mechanisms involved in pathogenesis of this disorder remain up to now unclear and are multifactorial. There are several factors that contribute to the hyperandrogenemia and anovulation in this condition.

A primary Neuroendocrine Defect Leading To Exaggerated LH
Pulse Frequency And Amplitude

A persistent finding in a majority of women with PCOS is abnormal gonadotropin secretion, particularly, elevated levels of LH. It was hypothesized that enhanced LH stimulation of the ovaries results in excess androgen secretion. This hypothesis was supported by studies using GnRH agonists which decreased serum LH, testosterone and androstendione, whereas DHEA-S and other adrenal androgens were unchained (Steingold et al., 1987). However, recent data from human and animal models suggests that the rapid GnRH pulse frequency is not a primary hypothalamic abnormality appreciating the effect of abnormal plasma levels of estrogen, insulin, or androgen (Poretsky et al., 1994; Dunaif et al., 1996; Dumesic et al., 1997).

A Defect Of Androgen Synthesis That Results In Enhanced
Ovarian Androgen Production


PCOS theca cells show increased activity of multiple steroidogenic enzymes such as 17-hydroxylase and 17,20-lyase, resulting in raised androgen production, both basally and in response to LH (Franks et al., 1999; Nelson et al., 2001). These abnormalities are presented in both theca cells in women with polycystic ovaries and chronic anovulation or women with regular ovulatory cycles and polycystic ovaries (Franks et al., 1999).

Recent human and animal studies provided several mechanisms by which a primary defect resulting in hyperandrogenemia could cause PCOS. First, the decreased sensitivity of the GnRH pulse generator appears to be a consequence of raised circulating androgen levels, because androgen receptor blockade with flutamid abolishes this defect in PCOS (Eagleson et al., 2000). Second, studies on rhesus monkeys suggest that prenatal androgen exposure produces many features characteristic of the PCOS phenotype, such as increased LH secretion, ovarian hyperandrogenism, central obesity and defective insulin secretion (Eisner et al., 2000). Third, permanent alterations in LH secretion were demonstrated in women who were exposed to excess androgens during in utero development, such as women with congenital adrenal hyperplasia or with neonatal androgen-secreting neoplasm (Rosenfield et al., 1999). Fourth, Dörner et al., (1998) observed an approximately fourfold increased prevalence of PCO in women born since 1955 in East Germany following the massive application of insecticide DDT. The DDT – metabolite o, p’-DDD is a strong inhibitor of 3β-hydxysteroid dehydrogenase, and DDT may induce 17,20 lyase activity, implying a possible connection between cases of PCOS in women born after 1955 and prenatal DDT exposure.

A Unique Defect In Insulin Action And Secretion

Peripheral insulin resistance (and associated hyperinsulinemia) plays a significant role for the pathogenesis of PCOS (Dunaif, 1997), and may be the primary abnormality in the aetiology of hormonal derangement. Insulin can directly stimulate testosterone synthesis in human theca cells (Nestler et al., 1998) and also contributes to increased adrenal secretion, in part by enhancing adrenal sensitivity to ACTH (Moghetti et al., 1996). Insulin decreases sex hormone-binding globulin production by the liver, subsequently increasing free serum testosterone (Nestler et al., 1991).


On the other hand, if insulin resistance and hyperinsulinemia have an important pathogenic role in PCOS, why are not all patients with hyperinsulinemia also hyperandrogenic, like many women with type 2 DM? Furthermore, how do ovaries appear to be insulin-responsive in an insulin-resistant state?

Conn et al., (2000) showed that although 82 % of women with type 2 DM had polycystic ovaries in ultrasound, only 52 % had hyperandrogenism and / or menstrual disturbances, suggesting that hyperinsulinemia alone is not sufficient for expression of this syndrome.

In another study in the group of Asian women, Rodin et al., (1998) reported that the effects of type 2 DM and polycystic ovaries on insulin sensitivity were independent, suggesting that these changes in insulin sensitivity involve different mechanisms. It is possible that the insulin resistance and the reproductive disturbances reflect separate genetic defects and that insulin resistance unmasks the syndrome in genetically susceptible women.

1.5 Long-Term Disease Risks


 The abnormal hormonal milieu characteristics of PCOS may predispose to several conditions, which include type 2 diabetes mellitus, hypertension, dyslipidemia, cardiovascular disease and some malignancies.

Type 2 Diabetes Mellitus

The association between hyperinsulinemia and hyperandrogenism was first described in 1980 by Burghen et al. and lead to the assumption that PCOS has associated metabolic risks. A recent analysis of glucose tolerance among a larger group of 249 women with PCOS aged 14 to 44 years indicated a 31% prevalence of impaired glucose tolerance and a 7% prevalence of type 2 diabetes mellitus, rates significantly higher than those seen in among normally cycling controls (Legro et al., 1999)

Cardiovascular Disease

Women with PCOS have higher cardiovascular risk than weight-matched controls with normal ovarian function (Wild, 2002) due to elevated androgen levels, body fat distribution and insulin resistance. A number of studies have shown that women with PCOS exhibit an abnormal lipoprotein profile characterized by raised concentrations of plasma triglycerids, marginally elevated low-density lipoprotein (LDL) cholesterol, and reduced high-density lipoprotein (HDL) cholesterol (Dejager et al., 2001: Pirwany et al., 2001). Furthermore, an increased hepatic lipase activity has been documented. A recent prospective study has linked menstrual irregularity, about 80% of which is due to PCOS, to increased risk of fatal coronary heart disease (Solomon et al., 2002).



The prevalence of endometrial cancer (EM) appears to be increased in young women with PCOS (Niwa et al., 2000; Wild et al., 2000). Pierpont and co-workers (1998) in the retrospective study of 786 PCOS women and 1060 weight-matched control women in United Kingdom showed that women with PCOS were not at significantly increased risk of mortality or morbidity from breast cancer but were at increased risk of endometrial cancer (Odds ratio 5.3). One possible mechanism is the associated high and unopposed level of estrogen; estrogen stimulation leads to endrometrial hyperplasia and subsequently to adenocarcinoma. Another theory suggests that hyperandrogenemia and hyperinsulinemia may increase the potential for neoplastic change in the endometrium through their effects on concentrations of SHBG, IGF-I and circulating estrogens (Meirow and Schenker, 1996).

The results concerning an association between PCOS and ovarian cancer are conflicting. One group (Coulam et al., 1983) showed no risk of ovarian carcinoma among anovulatory women, while others (Schildkraut et al.,1996) suggested a relative risk of 2.5 (95% CI 1.1–5.9) in the case-control study. In the large UK study the standardized mortality for ovarian cancer was 0.39 (95% CI 0.01-2.17).

1.6 Management of PCOS

Women with PCOS seek medical help in order to reduce a hair growth and/or acne, to restore a menstrual cyclicity and infertility. In addition, these patients are increasingly seeking treatment for the metabolic abnormalities such as insulin resistance and obesity. A “problem-oriented” approach to treatment of PCOS is the following:

Oral Contraceptives


Oral contraceptives are useful in patients with PCOS who do not desire pregnancy. Besides establishing regular menstrual cycle, the combined estrogen/progestin oral contraceptive pill inhibits endometrial proliferation and reduces ovarian androgen proliferation (Burkman RT, 1995). Hirsutism and acne respond well to oral contraceptive use. It is important to choose the appropriate oral contraceptive. Newer progestins such as desogestrel, as well as norgestimate and ethynodiol diacetate, have minimal androgenic potential and are considered to be superior to preparations, containing norgestrel or norethindrone, which have higher portal activity.


At least fifty percent or more of women with PCOS have a significant adrenal component to their hyperandrogenism, as evidenced by elevated concentrations of DHEAS. Such women generally benefit from the use of glucocorticoid preparation such as dexamethasone (0.5 mg/d) and prednisolone (5 mg/d). Low doses of these drugs provide a good suppression of adrenal androgen secretion without significant cortisol suppression.

Clomiphene Citrate

The antiestrogen clomiphene citrate (CC) remains the first-line medical therapy for ovulation induction in women with PCOS. The standard regimen is 50 mg/day for 5 days beginning on cycle day 5 following spontaneous or progestine-induced bleeding. The dose can be increased ( to a maximum dose of 250 mg/day) in subsequent cycles if serum progesterone in the luteal phase is less than 10 ng/ml. Clomiphene citrate induces ovulation in approximately 70% to 85% of patients, although only 33% to 45% conceive (Nasseri, 2001).


An improvement of the outcome in clomiphene citrate cycles is possible due to adjunct use of hCG. The dose of 5000 or 10000 U (intramuscularly) hCG triggers ovulation and results in LH surge in cases in which no spontaneous LH surge is detected.


The 15% of PCOS patients who fail clomiphene therapy are treated with gonadotropins.

The gonadotropin treatment of women with PCOS is relatively problematic because of high rates of multiple gestations and the occurrence of ovarian hyperstimulation syndrome. These problems can be avoided by use of low-dose, “step-up” regimens designed to result in the development of a single dominant follicle. The treatment starts at the dose of 37.5 U of FSH daily on day 3 after spontaneous or progestine–induced menses. The dose is increased every 7 days and an ovulatory hCG trigger is given when the lead follicle reaches a mean diameter of 18 mm. The “step-down” protocol involves higher starting FSH doses, followed by dose reduction when the leading follicle exceeds a mean diameter of 10 mm.

Ovarian Drilling


For patients who fail to respond to the use of injectable gonadotropin treatment, laparoscopic ovarian drilling is appropriate. The technique usually involves the laparoscopic cauterization of the ovarian surface (LCOS). A recent study of 1124 patients found spontaneous ovulation in 77% and pregnancy in 49% of women (Campo, 1998). In contrast to gonadotropins, the spontaneous abortion rate after a laparoscopic ovulation induction is only about 15% and multiple pregnancies are uncommon (2.5%). Even patients who fail to ovulate spontaneously after drilling may benefit due to improved response to gonadotropin treatment.

Given the importance of hyperinsulinemia in the pathogenesis of PCOS, it has been hypothesized that insulin-sensitizing agents may be useful in the restoration of normal endocrinological and metabolic parameters.

The most extensively studied insulin-sensitizing drug in the treatment of PCOS is metformin.



Metformin is thought to have primary effects on peripheral glucose uptake in response to insulin, with some reduction in basal hepatic glucose production (Mehnert, 2001). It also lowers adipose-tissue lipolysis and improves insulin sensitivity in muscle (Witters, 2001). Its mechanism of action is not defined but recent findings suggest a unifying role of AMP-activated protein kinase in all the mechanisms of metformin action (Zhou et al., 2002).The drug does not provoke hyperinsulinemia and therefore does not cause hypoglycaemia. It is now recommended as first-line therapy in overweight patients with diabetes by most leading clinical associations. It is also inexpensive.

Non-Randomised Studies With Metformin

In PCOS the first pioneer study with metformin was conducted 1994 by Velasquez and co-workers. Most of following trials had cohort designs and showed an improvement in insulin metabolism and a reduction in circulating androgen concentrations (Crave et al., 1995; Ehrmann et al., 1997; Morin-Papunen et al., 1998; Glueck et al., 1999). In most cases, small reductions were seen in body-mass index, waist/hip ratio, or both, and improvements in menstrual cyclicity (presumed ovulation) were also found. Only one of these trials (Crave et al., 1995) examined the effect of metformin on hirsutism, and there was no reported evidence on acne. In general, the results were encouraging, but all trials involved small numbers of patients, and most were of short duration and limited in design by not having a control or placebo group.

Concerning a menstrual cycle, the range of benefits in uncontrolled studies is wide. The normal menstrual frequency was achieved in 16% (four of 24 cases) ( Crave et al., 1995) and in more than 90% (39 of 43 cases) (Glueck et al., 1999) in women with PCOS.

Results From Controlled Studies


There have been seven published studies on metformin that included some form of randomisation (control group with or without placebo), five of them were double-blind in design ( Nestler et al., 1998; Moghetti et al., 2000; Morin-Papunen et al., 2000; Pasquali et al., 2000; Ng et al., 2001; Fleming et al., 2002; Kocak et al., 2002). The most consistent findings, were a decrease in body-mass index of around 4% and in androgen measures of around 20%, compared with placebo. The data on improvements in insulin concentrations and, in particular, SHBG are less convincing when considered together with placebo data. These observations show the potential for confounding effects during any prospective studies and re-emphasise the importance of control in study design.

For ovulation, the most important observations were that the interval from start of treatment to first ovulation was significantly shorter with metformin than with placebo, that menstrual or ovulation cyclicity was increased with metformin, and that these improvements were variable and modest.

Taken together, studies with metformin have controversial results concerning metabolic changes, androgen levels, ovulatory function and pregnancy. There were many variations in the patients examined and the methods of assessments used. Several points such as metformin dose and its relation to body mass remains unclear.

1.7 Aim Of The Trial


Therefore, the present randomised, double blind, placebo-controlled study was designed to compare the clinical, endocrine and metabolic effects of two treatment models for PCOS:

1) Metformin and lifestyle modification, including behavioural group therapy with aspects in nutrition and physical activity, and individual counselling by dietician.

2) Placebo and lifestyle modification, including behavioural group therapy with aspects in nutrition and physical activity, and individual counselling by dietician.


Changes in endocrinological and metabolic parameters such as FSH, LH, prolactin, estradiol, testosterone, free testosterone, DHEA-S and SHBG, total cholesterol, triglycerides, HDL-cholesterol, LDL-cholesterol, insulin, glucose, HbA1c, leptin and IGF-I should be assessed before and after therapy. Modification in insulin and glucose levels should be evaluated by means of AUC (area under the curve).

The second purpose of this trial was to reveal an influence of metformin on the menstrual cycles and spontaneous pregnancy.

Forty six caucasian women with PCOS aged 21 to 39 were selected in the reproductive medicine department at the University Clinic Charité, Berlin, Germany.


The patients were divided into two groups. The first group received metformin and the second received a placebo during the sixteen weeks of the trial.

All laboratory investigations: FSH, LH, prolactin, progesterone, estradiol, testosterone, free testosterone, DHEA-S and SHBG, total cholesterol, triglycerides, HDL-cholesterol, LDL-cholesterol, HbA1c, leptin and oral glucose tolerance test were performed every 8 weeks.

Body composition was studied by the bioelectrical impedance method at week 0, 8 and 16. Based on Resistance ® , Reactance (Xc) and the phase angle alpha measured at 5, 50 and 100 kHz body compartments (total body water, fat-free mass, body cell mass and fat mass) were calculated.


This study was approved by the Charité ethic commission, Berlin, Germany .

© Die inhaltliche Zusammenstellung und Aufmachung dieser Publikation sowie die elektronische Verarbeitung sind urheberrechtlich geschützt. Jede Verwertung, die nicht ausdrücklich vom Urheberrechtsgesetz zugelassen ist, bedarf der vorherigen Zustimmung. Das gilt insbesondere für die Vervielfältigung, die Bearbeitung und Einspeicherung und Verarbeitung in elektronische Systeme.
DiML DTD Version 4.0Zertifizierter Dokumentenserver
der Humboldt-Universität zu Berlin
HTML generated: